Pancreatic cancer is associated with an extremely poor prognosis. Even after resection with curative intent, the 5-year survival rates usually do not exceed 5%, and neither radiotherapy (RT) nor chemotherapy (CT) improve long term survival. Because of the poor outcome of patients with localized disease, studies should be directed to selected patients who benefit from surgery and from combined treatment of surgery with chemo-irradiation. Future developments in the treatment of locoregionally confined pancreatic cancer will have to focus on treatment modalities which improve the local tumor control as well as the control of occult distant disease. Combined CT/RT seems to be the most promising approach for these patients. However, to date there are more questions than clearcut answers about what such a multimodality treatment should look like. This concerns the selection of patients, the dosages, time schedules, and routes of administration of CT and RT. Last but not least, there is a need for more effective CT in pancreatic cancer. New drugs and multiple drug combinations should be studied in patients who are not candidates for multimodality treatment because of distant metastases. Well designed trials that address these issues are urgently needed in the forthcoming years.